OECD MCP Server
Provides AI assistants access to over 5,000 OECD economic and statistical datasets via the SDMX API for search, analysis, and comparison across 38 countries.
README
<p align="center"> <img width="640" height="320" alt="OECD MCP Server banner" src="https://github.com/user-attachments/assets/25c886ab-4857-4960-be26-dd250f3155b0" /> </p>
OECD MCP Server
A Model Context Protocol (MCP) server providing AI assistants access to OECD's 5,000+ economic and statistical datasets via SDMX API. By connecting to the MCP server, you can use AI to search, analyze, and compare data from 38 OECD countries across economy, health, education, environment, and more.
<details> <summary>🇸🇪 Svenska</summary>
En Model Context Protocol (MCP) server som ger AI-assistenter tillgång till OECD:s 5,000+ ekonomiska och statistiska dataset via SDMX API. Genom att ansluta till MCP-servern kan du med hjälp av AI söka, analysera och jämföra data från 38 OECD-länder inom ekonomi, hälsa, utbildning, miljö och mer.
</details>
Quick Start
<details> <summary>🇸🇪 Snabbstart (Svenska)</summary>
Det är enkelt att ansluta en LLM / AI-chatbot till MCP-servern. Det enklaste sättet är att använda npm-paketet oecd-mcp via npx. Nedan finner du guider för olika klienter.
</details>
The easiest way to connect an LLM / AI chatbot to the MCP server is via the published npm package oecd-mcp using npx. Below you'll find guides for different clients.
Claude Desktop
1. Open Claude Desktop settings → Developer → "Edit Config"
2. Add to the JSON file:
{
"mcpServers": {
"oecd": {
"command": "npx",
"args": ["-y", "oecd-mcp"]
}
}
}
3. Save and restart Claude Desktop
Claude Code
claude mcp add oecd npx -y oecd-mcp
Verify: claude mcp list
OpenAI Codex
~/.codex/config.toml:
[mcp.oecd]
command = "npx"
args = ["-y", "oecd-mcp"]
transport = "stdio"
From Source
git clone https://github.com/isakskogstad/OECD-MCP.git
cd OECD-MCP
npm install
npm run build
Then point your client at node /absolute/path/to/OECD-MCP/dist/index.js, e.g.:
{
"mcpServers": {
"oecd": {
"command": "node",
"args": ["/absolute/path/to/OECD-MCP/dist/index.js"]
}
}
}
Features
<details> <summary>🇸🇪 Funktioner (Svenska)</summary>
Servern ansluter till OECD SDMX API och ger tillgång till 5,000+ dataset i 17 kategorier.
OECD SDMX API Tillgång till kvartalsvis nationalräkenskaper (QNA), huvudsakliga ekonomiska indikatorer (MEI), hälsostatistik, utbildningsdata, miljöindikatorer och mer från 38 OECD-medlemsländer plus partnerekonomier.
MCP-servern implementerar MCP-protokollet med stöd för:
- 9 verktyg — Dataset-upptäckt (5) och dataåtkomst (4)
- 7 resurser — Kategorier, länder, filterguide, ordlista, LLM-instruktioner
- 7 promptmallar — Ekonomisk analys, landsjämförelser, nordiskt fokus
</details>
The server connects to the OECD SDMX API providing access to 5,000+ datasets across 17 categories:
OECD SDMX API Access to quarterly national accounts (QNA), main economic indicators (MEI), health statistics, education data, environmental indicators, and more from 38 OECD member countries plus partner economies.
The MCP server implements the MCP protocol with support for:
- 9 tools — Dataset discovery (5) and data access (4)
- 7 resources — Categories, countries, filter guide, glossary, LLM instructions
- 7 prompt templates — Economic analysis, country comparisons, Nordic focus
Tools
| Tool | Description |
|---|---|
search_dataflows |
Search for datasets by keyword |
list_dataflows |
List datasets by category |
get_categories |
List all 17 data categories |
get_popular_datasets |
Get commonly used datasets |
search_indicators |
Search for specific indicators |
get_data_structure |
Get metadata for a dataset |
query_data |
Query statistical data |
get_dataflow_url |
Generate OECD Data Explorer link |
list_categories_detailed |
Detailed category information |
Resources
| Resource | Description |
|---|---|
oecd://categories |
17 data categories with descriptions |
oecd://dataflows/popular |
Curated popular datasets |
oecd://countries |
ISO 3166-1 alpha-3 country codes |
oecd://filter-guide |
SDMX filter syntax guide |
oecd://glossary |
Definitions of OECD terms |
oecd://llm-instructions |
Instructions for AI assistants |
oecd://api/info |
API information |
Prompt Templates
| Prompt | Description |
|---|---|
analyze_economic_trend |
Analyze economic trends over time |
compare_countries |
Compare data between countries |
get_latest_statistics |
Get latest statistics |
explore_dataset |
Guided dataset exploration |
find_data_for_question |
Find the right dataset for a question |
build_filter |
Help building SDMX filters |
nordic_comparison |
Compare Nordic countries |
Use Cases
<details> <summary>🇸🇪 Användningsområden (Svenska)</summary>
Ekonomer & Forskare
Verktyg:
| Verktyg | Beskrivning |
|---|---|
query_data |
Fråga BNP, inflation, arbetslöshetsdata |
search_dataflows |
Hitta dataset för forskningsfrågor |
get_data_structure |
Förstå dataset-dimensioner |
Exempel:
- "Jämför BNP-tillväxt mellan G7-länderna 2020-2024"
- "Analysera inflationstrender i eurozonen"
- "Hitta data om arbetsproduktivitet per sektor"
App-exempel: Mobil ekonomisk brief med BNP-trender och jämförelsediagram.
Policyanalytiker
Verktyg:
| Verktyg | Beskrivning |
|---|---|
get_categories |
Bläddra i 17 policyområden |
list_dataflows |
Hitta relevanta policy-dataset |
get_dataflow_url |
Generera länkar för rapporter |
Exempel:
- "Vilken sjukvårdsutgiftsdata finns för Sverige?"
- "Hitta utbildningsresultatindikatorer för PISA-analys"
- "Jämför miljöpolicyer i nordiska länder"
App-exempel: Hälsostatistik-dashboard med donutdiagram och landjämförelser.
Datajournalister
Verktyg:
| Verktyg | Beskrivning |
|---|---|
search_indicators |
Hitta specifika mätvärden |
get_popular_datasets |
Få tillgång till ofta citerad data |
query_data |
Hämta siffror för artiklar |
Exempel:
- "Hämta senaste arbetslöshetssiffror för EU-länder"
- "Hitta inkomstojämlikhetsdata (Gini-koefficient)"
- "Jämför skatteintäkter som % av BNP"
App-exempel: Produktivitetsjämförelse med stapeldiagram och trendlinjer.
</details>
Economists & Researchers
Tools:
| Tool | Description |
|---|---|
query_data |
Query GDP, inflation, unemployment data |
search_dataflows |
Find datasets for research questions |
get_data_structure |
Understand dataset dimensions |
Examples:
- "Compare GDP growth between G7 countries 2020-2024"
- "Analyze inflation trends in the Eurozone"
- "Find data on labor productivity by sector"
App Example:
<img width="320" alt="GDP Economic Brief" src="https://raw.githubusercontent.com/isakskogstad/OECD-MCP/main/assets/app-examples/app1-mobile-minimalist.png" />
Mobile economic brief with GDP trends, G7 comparison charts, and sparkline visualizations from the QNA dataset.
Policy Analysts
Tools:
| Tool | Description |
|---|---|
get_categories |
Browse 17 policy areas |
list_dataflows |
Find relevant policy datasets |
get_dataflow_url |
Generate links for reports |
Examples:
- "What healthcare spending data is available for Sweden?"
- "Find education outcome indicators for PISA analysis"
- "Compare environmental policies across Nordic countries"
App Example:
<img width="500" alt="Health Statistics" src="https://raw.githubusercontent.com/isakskogstad/OECD-MCP/main/assets/app-examples/app2-light-minimalist.png" />
Health statistics dashboard with donut chart overview and country comparison from the HEALTH_STAT dataset.
Data Journalists
Tools:
| Tool | Description |
|---|---|
search_indicators |
Find specific metrics |
get_popular_datasets |
Access commonly cited data |
query_data |
Get numbers for stories |
Examples:
- "Get latest unemployment figures for EU countries"
- "Find income inequality data (Gini coefficient)"
- "Compare tax revenues as % of GDP"
Data Categories
<details> <summary>View all 17 categories</summary>
| Category | Code | Description | Datasets |
|---|---|---|---|
| Economy | ECO | GDP, growth, inflation, interest rates | 500+ |
| Employment | JOB | Labor market, wages | 300+ |
| Trade | TRD | International trade | 250+ |
| Health | HEA | Healthcare, life expectancy | 200+ |
| Education | EDU | PISA, outcomes | 180+ |
| Environment | ENV | Climate, emissions | 150+ |
| Innovation | STI | R&D, patents, AI | 140+ |
| Energy | NRG | Production, renewables | 120+ |
| Taxation | TAX | Tax revenues, tax rates | 100+ |
| Finance | FIN | Markets, banking | 95+ |
| Government | GOV | Governance, public sector | 90+ |
| Social | SOC | Inequality, quality of life | 85+ |
| Agriculture | AGR | Production, food security | 75+ |
| Industry | IND | Industrial production | 70+ |
| Development | DEV | Development aid | 60+ |
| Transport | TRA | Infrastructure, mobility | 50+ |
| Regional | REG | Subnational data | 45+ |
</details>
About
<details> <summary>🇸🇪 Om (Svenska)</summary>
Skapad av: Isak Skogstad
Kontakt: isak.skogstad@me.com
Fristående: Detta projekt är fristående och är inte anslutet till, godkänt av eller officiellt kopplat till OECD (Organisationen för ekonomiskt samarbete och utveckling).
Användarvillkor: Data tillhandahålls via OECD:s öppna SDMX API. Se OECD:s användarvillkor för användningsvillkor.
</details>
Created by: Isak Skogstad
Contact: isak.skogstad@me.com
Disclaimer: This project is independent and not affiliated with, endorsed by, or officially connected to the OECD (Organisation for Economic Co-operation and Development).
Terms of use: Data is provided via OECD's open SDMX API. See OECD Terms and Conditions for usage conditions.
Resources
- OECD Data Portal: https://data.oecd.org/
- SDMX Standard: https://sdmx.org/
- MCP Documentation: https://modelcontextprotocol.io/
- npm Package: https://www.npmjs.com/package/oecd-mcp
- Issues: GitHub Issues
Built with Model Context Protocol SDK | Version 4.0.0
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。